Robert Y. Hou

Disk arrays: high-performance, high-reliability storage subsystems

As the performance of other system components continues to improve rapidly, storage subsystem performance becomes increasingly important. Storage subsystem performance and reliability can be enhanced by logically grouping multiple disk drives into disk arrays. Array data organizations are defined by their data distribution schemes and redundancy mechanisms. The various combinations of these two components make disk arrays suitable for a wide range of environments. Many array implementation decisions also result in trade-offs between performance and reliability. Disk arrays are thus an essential tool for satisfying storage performance and reliability requirements, while proper selection of a data organization can tailor an array to a particular environment.<<ETX>>

Balancing I/O response time and disk rebuild time in a RAID5 disk array

When a disk in the RAID5 disk array architecture has failed, requests to that disk can only be serviced by reading data from all surviving disks and rebuilding the lost data. This may cut disk performance in half. To avoid this degradation, all of the lost data must be rebuilt and written to a spare disk. The faster the data are rebuilt, the sooner the disk array returns to normal operation. Giving high priority to the rebuild process, however, can increase response times for incoming application requests which complete for disk service. A balance must be found between acceptable application response times and disk rebuild times. Simulation was used to evaluate the effect of the rebuild unit size on response time and rebuild time. The authors have found this tradeoff to be embodied in the choice of the rebuild unit and the amount of rebuild data which is atomically read from each surviving disk. The find that a single track rebuild unit provides faster rebuild times than a one sector rebuild unit. Rebuilding one track at a time provides better application request response times when compared with rebuilding one cylinder at a time.<<ETX>>

Comparing rebuild algorithms for mirrored and RAID5 disk arrays

Several disk array architectures have been proposed to provide high throughput for transaction processing applications. When a single disk in a redundant array fails, the array continues to operate, albeit in a degraded mode with a corresponding reduction in performance. In addition, the lost data must be rebuilt to a spare disk in a timely manner to reduce the probability of permanent data loss. Several researchers have proposed and examined algorithms for rebuilding the failed disk in a disk array with parity. We examine the use of these algorithms to rebuild a mirrored disk array and compare the rebuild time and performance of the RAID5 and mirrored arrays. Redirection of Reads provides comparable average response times and better rebuild times than Piggybacking for a mirrored array, whereas these two algorithms perform similarly for a RAID5 array. In our experiments comparing the two architectures, a mirrored array has more disks than a RAID5 array and can sustain 150% more I/Os per second during the rebuild process. Even if the size of the RAID5 array is increased to match the mirrored array, the mirrored array reduces response times by up to 60% and rebuild times by up to 45%.

Using non-volatile storage to improve the reliability of RAID5 disk arrays

RAID5 disk arrays are becoming popular solutions for providing fast access to data for transaction processing applications. They provide good performance at low cost without sacrificing much data reliability. Their main drawback is poor performance for small write requests. Techniques for using non-volatile RAM (NVRAM) have been proposed for improving the performance of these write requests. The paper shows that NVRAM and other nonvolatile devices can also be used to improve the reliability of an array by significantly reducing the time required to repair a failed disk: in the array in the event of a single disk failure. More importantly, proper use of these non-volatile devices allows the array to support heavier workloads than previously reported yet still repair a failed disk in a reasonable amount of time.

Trading disk capacity for performance

Improvements in disk access time have lagged behind improvements in microprocessor and main memory speeds. This disparity has made the storage subsystem a major bottleneck for many applications. Disk arrays that can service multiple disk requests simultaneously are being used to satisfy increasing throughput requirements. Higher throughput rates can be achieved by increasing the number of disks in an array. This increases the number of actuators that are available to service separate requests. It also spreads the data among more disk drives, reducing the seek time as the number of cylinders utilized on each disk drive decreases. The result is an increase in throughput that exceeds the increase in the number of disks. This suggests a tradeoff between the space utilization of disks in an array and the throughput of the array.<<ETX>>